NASA TN D-337
CENTRIFUGE STUDY OF PILOT TOLERANCE TO ACCELERATION
AND THE EFFECTS OF ACCELERATION ON
By Brent Y. Creer, Captain Harald A. Smedal, USN (MC),
and Rodney C. Wingrove
Ames Research Center
Moffett Field, Calif.
NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
WASHINGTON November 1960
IN NATIONAL AERONAUTICS AND SPACE ADMINISTRATION
TECHNICAL NOTE D-337
CENTRIFUGE STUDY OF PILOT TOLERANCE TO ACCELERATION
AND THE EFFECTS OF ACCELERATION ON
By Brent Y. Creer, Captain Harald A. Smedal, USN (MC),
and Rodney C Wingrove
A research program, the general objective of which was to measure
the effects of various sustained accelerations on the control performance
of pilots, was carried out on the Aviation Medical Acceleration Laboratory
centrifiige, U. S. Naval Air Development Center, Johns v llle , Pa. The
experimental setup consisted of a flight simulator with the centrifuge
in the control loop. The pilot performed his control tasks while being
subjected to acceleration fields such as might be encountered by a
forward -facing pilot flying an atmosphere entry vehicle. The study was
divided into three phases •
In one phase of the program, the pilots were subjected to a variety
of sustained linear acceleration forces while controlling vehicles with
several different sets of longitudinal dynamics- Here, a randomly
moving target was displayed to the pilot on a cathode -ray tube. For
each combination of acceleration field and vehicle dynamics, pilot track-
ing accuracy was measured and pilot opinion of the stability and control
characteristics was recorded. ThuS; information was obtained on the
combined effects of complexity of control task and magnitude and direction
of acceleration forces on pilot performance . These tests showed that
the pilot's tracking performance deteriorated markedly at accelerations
greater than about ^g when controlling a lightly damped vehicle . The
tentative conclusion was also reached that regardless of the airframe
dynamics involved, the pilot feels that in order to have the same level
of control over the vehicle , an increase in the vehicle dynamic stability
was required with increases in the magnitudes of the acceleration
impressed upon the pilot.
In another phase, boundaries of human tolerance of acceleration
were established for acceleration fields such as might be encountered by
a pilot flying an orbital vehicle. A special pilot restraint system
was developed to increase human tolerance to longitudinal decelerations.
The results of the tests showed that human tolerance of longitudinal
deceleration forces v . considerably improved through use of the special
A conparative evaluation was made, iji aiother phase of the Invest i-
i-atlon, of the three-axis type of side-arm controller and the two-axis
type in combination with toe pedals for yaw control. During the tests,
the difficulty of blending and applying three control -inputs with one
hand using the three -axis controllers was repeatedly pointed out hy the
evaluation pilots; as a result, they were ur.animous in their preference
of the two-axis toe-pedal class of controllers.
There have "been numerous research investigations conducted on the
effects of acceleration forces on man. These experments were focused
principally upon the medical aspects of man's tolerance to acceleration
forces with only secondary interest in asset's ing the influence of
acceleration forces on the human's ability 1,o perform a task (refs. 1
through 13). The results of these research studies have heen of great
value in the initial design studies of man-carrying orbital vehicles.
However, it appears that man will eventually- be called upon to assume
manual control of an orbital vehicle. This may come ahout because of
a failure in the automatic control system o::- it may be a routine piloting
task. It appears, therefore, that much more information is needed on
the influence of acceleration on man's ability to perform a complex
In addition, most of the studies on mai's tolerance to sustained
accelerations were made using nonpilot test subjects. It is probable
that onl^ highly motivated test pilots will be used to man the orbital
or near orbital vehicles. The fairly large differences in time tolerance
to accelerat-ion for pilot and nonpilot subj ;cts were demonstrated in
reference 12. It is generally accepted tha: the pilot's performance
in and tolerance to acceleration fields are critically dependent upon
the pilot's restraint system. The restrain:: systems used in many of
the past studies were of course not represe itative of the current state
of the art. It would therefore appear that additional tests are
required, using test pilot subjects and representative restraint systems,
to define pilot tolerance to sustained accelerations.
Recent work conducted by the National Aeronautics and Space
Adjninistration was focused directly on the problems of a pilot flying
a vehicle during launch, or along an atmosjtiere entry trajectory
(refs. 1^1- through l6). In these studies tie principal objective was
assessing the pilot's ability to control tie vehicle while flying in an
elevated'g field. However, these studies vere rather specific in nature.
As part of the general NASA program, e. study was conducted by the
Ajiies Research Center (during Sept. 19^9) or. the Aviation Medical Accelera-
tion Laboratory centrifuge. Naval Air DeveD.opment Center, Johnsville , Pa.
For this experiment^ which was fairly 'jeneral, the flight simulator
experimental setup utilized the centrifuge in the control loop. The
subject pilots were seated in the gondola of the centrifuge and were
confronted with a fairly complex task which involved flying a simulated
orbital vehicle entering the atraosphere. This study was split into three
phases. The objectives of each phase were as follows:
(1) To obtain information on the combined effects of magnitude and
direction of the applied acceleration force and of control task complexity
on the pilot ' s performance .
(2) To establish some meaningful tolerance to acceleration times
for the direction of acceleration fields encoimtered by a pilot in a
forward-facing position flying along an atmosphere entry trajectory.
A special anterior restraint system was developed in an attempt to
increase human tolerance to longitudinal decelerations. Time tolerance
to acceleration runs were also made for other directions of acceleration
(3) A preliminary centrifuge investigation was conducted wherein
several side-arm controllers were evaluated. One objective was to compare
three-axis controllers with the two-axis, toe -pedal-type airplane controls.
Tlie toe -pedal-type control used was designed to minimize the effects of
acceleration on the pilot's yaw control inputs.
Tills study was brief and of an exploratory nature. Nevertheless,
it is believed that the results will be of value to the orbital-vehicle
design engineer. In this paper, the vernacular of the test pilot has
been used to describe the direction of the applied acceleration force.
Tlie terms "eyeballs in/' "eyeballs out," and "eyeballs down" correspond
to acceleration fields Ax, -Ax, and Ajj, respectively, where Ax, -Ax,
and Aj,] refer to the direction of acceleration forces measured in the
conventional airplane body-axis coordinate system.
Ajj acceleration factor, ratio of acceleration force to weight,
positive when directed upward along spinal axis (i.e., from
seat to head)
Ax acceleration factor, ratio of acceleration force to weight,
positive when directed forward transverse to spinal axis
(i.e., from back to chest)
c wing reference chord, ft
Cmc. variation of pitching -moment coefficient with Sp . per radian
acceleration of gravity, ft/sec^
ly moment of Inertia about vehicle Y axis^ slug-ft^
Msg ^ ^^^5^^ per sec^
q dynamic pressure, Ib/sq ft
S reference wing area, sq ft
5e elevator deflection, radians
5p pilot stick deflection, deg 1^
^ damping ratio of longitudinal oscillatory mode of motion ^
0Ji:t natural frequency of longitudinal oscjllatory mode of motion,
APPARAIUS MD TEST PROCEDURES
With regard to apparatus used in this tesi , the centrifuge at
Jolinsville, Pa-, has received extensive coverage in the nation's
magazines and technical journals and it will be assumed that everyone
is f-;enerally familiar with this device- For a fairly detailed descrip-
tion of the centrifuge see references 17 and 1^ .
The pilot's restraint system used in the centrifuge tests is shown
in figure 1- For protection against eyeballs-in accelerations it was
felt that a pilot's couch similar to the type xsed. in the Project
Mercury capsule would be adequate for this stucy. Individual molds were
made for each pilot. In figure 1, the pilot we.s essentially In a sitting
position, with his upper body and head held at an angle of 850 to 90°
with reference to the thigh position. The lower end of the leg mold in
the vicinity of the ankles and the feet was cui off to permit the
installation of the toe pedals for yaw control The pilot's feet were
restrained by strapping them in the toe -pedal c.evices. It might be
noted that the toe pedals were actuated by differential rotation of the
feet about the ankle joint- Thus, no movement of the leg was required
and the entire leg could be firmly restrained. The head restraint, which
is a critical item for eyeballs-out accelerations, was incorporated in
the helmet system. The helmet was secured into the mold by nylon straps
which were attached on each side of the helmet- Face pieces, which were
used to restrain the head in the helmet , were '.ndividually molded from
plaster cast impressions of each pilot's face- They were designed so
thai: the major portion of the load would be taien over the prominences
of the malar bones of the face- The chin cup -ras included in this
restraint system, "but only as a minor component since the chin is an
unstable support point and its tolerance to large loadings is poor.
Tlie face plates were attached to the helmet by adjustable nylon straps
fitted into a standard oxygen mask assembly.
The upper half of the torso was restrained by a bib fabricated of
straps crossed over the upper portion of the chest so that most of the
loading was taken over the upper rib cage- The rather snug fitting bib
restricted the expansion of the upper chest. Therefore, the frontal
area over the abdomen and lower chest was left essentially unsupported
to allow excursion of the diaphragm and movement of the lower rib cage
during the normal breathing process. Another separate component was
fabricated for the pelvis. This consisted of two slightly crossed
straps which were positioned to carry the loading over the pelvic bones
and the upper thighs .
The liJTib restraints were constructed of nylon netting. All anterior
restraints were extended through the mold and secured to the structure
which supported the styrofoam couches. A more detailed description of
the pilot's restraint system is given in reference 19- It should be
noted that anti-g suits were worn by all test subjects.
The pilots instrument display is shown in figure 2- A cathode-ray
tube in the instriiment panel was used to display a randomly driven
doughnut -shaped target. The dashed line on the display was drawn to
illustrate that the target motion always remained on a line which passed
through the center of the airplane reference and was perpendicular to
the horizon. The vehicle roll and pitch attitude were displayed on the
scope in the same fashion as they appear on a normal gyro horizon
indicator. The sideslip angle was presented on the scope by the lateral
displacement of the short vertical line away from the center index.
For all phases of the investigation, except the evaluation phase
of the side-arm controllers, the pilot controls consisted of a finger
operated two-axis side-arm controller and toe pedals. A description of
the finger operated side-arm controller and of the toe -pedal controls
is given in the last section of this report .
With regard to test conditions and procedures, the pilot flew the
centrifuge as a closed-loop system; that is , for acceleration fields
greater than 1 g, the centrifuge was driven in response to the pilot
control inputs in such a fashion that the impressed linear accelera-
tions varied in the same manner as the linear accelerations computed
from the aircraft equations of motion. A detailed description of the
closed-loop centrifuge operation is given in reference ik. The test
setup was arranged so that the total g field impressed on the pilot
consisted of two separate components; to a specified constant (biased) g
field was added the computed perturbations in normal and side accelera-
tion which resulted from the vehicle maneuvering about a given trim
condition. The pertiirbations in side and normal accelerations were
generally not greater than ±0.5g- In this experiment, the aircraft
equations of motion described five degrees of freedom with the vehicle
forward velocity assumed constant.
EFFECTS OF ACCELERATION AND COMTROL TASK ON PILOT PERFORMANCE
In this phase of the experiment, six different acceleration fields
were investigated. The maximum accelerations investigated were 6g in
an eyeballs-in direction, 6g in an eyeballs-down direction, and 7g in
an eyeballs -out direction. A number of runs were made in each accelera- -^
tion field with the complexity of the control task as the variable. The ^
complexity of the control task was varied by charsging the damping and 3
frequency of the vehicle longitudinal short-pericd oscillation. The o
dynamic characteristics of the roll and yaw modes of airframe motion
were held constant. Table I presents the lateraj -directional and the
longitudinal airframe dynamics used in this phase: of the study.
A qualitative measure of pilot performance was obtained by having
the pilot give a numerical rating on the control] ability of the simu-
lated vehicle by using the pilot opinion rating schedule presented in
table II. This pilot opinion schedule is essentially that presented in
reference 20. In order to obtain a quantitative measure of the pilot's
performance, a tracking task was utilized. The jiilot's tracking score,
which was the quantitative index of pilot perforr.ance , was calculated
as the accumulated tracking error compared with 1 he accumulated excursions
of the target as expressed in the following equaiion:
T p T
Qi^dt - / e^dt
Pilot tracking score = — ~ ' °
0j_2 the square of the target excursions
e2 the square of the tracking error excursions
T time interval of the tracking task
A detailed description of this tracking task is ijresented in reference
21. The length of the centrifuge tasks was 2-l/;:; minutes. Approximately
I-I/2 minutes were devoted to the pilot's assessing the controllability
of the system, the last minute being devoted to the tracking task. It
might be noted that during the latter part of the 1- minute tracking task.
the integrated pilot tracking score was still fairly sensitive to the
pilot's instantaneous tracking error.
Figure 3 presents the tracking scores obtained from these tests
for one of the subject pilots . This particular pilot was experienced
in ridmg the centrifuge and was thoroughly familiar with the tracking-
task, and the data obtained from his test runs were believed to be
representative of a well-trained pilot preconditioned to the oi-rgcts
of acceleration forces. His tracking score is plotted against the
magnitude of the g force. Data for the eyeballs -down , eyeballs -out
A and eyeballs -in accelerations are given for well-dajn-ped vehicle motions
4 aJ^cl for lightly damped vehicle motions. The well-damped case corresponde
3 to a fairly easy control task and the lightly damped case corresponds
6 to a fairly difficult control task. Certain tentative conclusions may
be drawn from these data. To a first approximation, it appears that
any decrement in pilot's tracking score is independent of the direction
of the applied acceleration investigated in this program. Pilot's
tracking score deteriorated markedly at accelerations greater than about
^g for the lightly damped dynamic situation. Finally, it appears that
the more difficult cont-ol task greatly magnifies any deficiencies in
the pilot's performance.
_The results of the pilot's ratings on the longitudinal handling
qualities of the vehicle obtained from these same perforaiance runs are
shown m figure 4. Pilot opinion boundaries which define satisfactory
unsatisfactory, and unacceptable regions of controllability of an entrv
vehicle are shown in terms of the period and damping ratio of the longi-
tudinal oscillatory mode of motion. The pilot ratings which defined the
various boundaries have been labeled in figure h and were as follows :
satisfactory-unsatisfactory = pilot rating 3-I/2
unsatisfactory-unacceptable = pilot rating 6-I/2
A curve corresponding to a pilot rating of 5 has been included since
this boundary defines the region of "unacceptable for normal operation."
The solid-lme boundaries to the left of the shaded regions were derived
from a moving cockpit flight simulator investigation (see ref . 22)
wherein the pilots were exposed to the earth's constant gravitational
field. The dashed-line boundaries to the right of the shaded regions
were obtained from the centrifuge tests wherein the pilots were immersed
in acceleration fields of approximately 6g to Tg- Thus, an Increase in
the acceleration field results in a corresponding shift in the pilot-
'^^^^^°\'^°^^^^^^^- This shift is from the solid-line boundary toward
the dashed-line boundary. The tentative conclusion is reached that
regardless of the region of airframe dynamics involved, the pilot feels
that m order to have the same level of control over the vehicle an
increase in the longitudinal dynamic stability, as shown by the shaded
area, is required with increases in the magnitudes of the acceleration
impressed upon the pilot. There is some logic to the above results.
The pilots often noted that more physical effort was required to control
manual dexterity and visual acuity may result with increases
accelerations impressed upon the pilot.
TIME TOtEEMCE TO ACCELERi^.TION
In the study to establish some meantagfu:. tolerance to acceleration
^ . ^ ,Jl\Tlr^+ of airframe dynamics was used. A description of
tS::V:hirif dyJa^ics is S::: ifta^le I. ^he pilot was faced with a
fairly difficulf task when controning this set °f/f -^-,- , ^^/^ ,
mSSudes of the accelerations investigated ranged from 6g to d-1/2 g
S the directions of the accelerations investigated were eyeballs in
Sehalls SL, and eyeballs out; a diagonal --^^^f^f ^ -^^eySaS -
investigated which consisted of a combination eyeballs-out and eyeballs
During the tolerance runs the pilot was required to fly the simu-
lated airplane and to the best of his ability, track the randomly _
lated airplane ^^, instructed to terminate the run if bodily pam
Sr :S::;ve'\rL^«^e .o f.t^,^. U^t ^e couM no longer co.
^f^ ZIoTA nature occurred. The project medical doctor monitored
tL"l™t°relectro=ardiogram and respirator,- recordings sn^ter^<.te^
Titir^aJ^iS^^ror^a "^ss-^s^-^-""- P--
:L?e"d^t:rSaSd mar.edl. A ti^ Msto^_ o. a t.pical^.^.al.^-out^
srpSrent^s s sirfifur^; -^^> - "^-rth^^^^fr^rirn't^Se
Pilot's elevator deflection, and a recording of the acceleration trace.
Se be Jiniing point for measuring tolerance :ime was taken when the
acceleration value was within about 10 perceit of that desired. It can
be seen from figure 5 that after the initial starting transients m
tracking score have subsided, the pilot's tracking efficiency remained
SriL^ly constant during the -^^^^ ^ -:-;,-- /.f:-::S;.
his vision deteriorated as the run progressed.
A brief survey or existing data on tim.to^ra^^^^^^^ sustained
re:S:r:nLTre:e1rinv^:rig^at1-in an .ttempt - ar-e at to^^^^^^^
bo^dariefo? time tolerance to acceleration are shown for comparison with
thfnewly established boundaries. A brief description is given of the
test coMitions procedures, and pilot's restraint system for each
ex^ri^ent ihicA contributed data on time tolerance to acceleration
to give the reader some Insight on the degree of confidence that can
be placed m the new proposed tolerance boundaries. In addition the
presentation of this information should provide the reader with a
better understanding of the differences between the currently accepted
and the proposed tolerance to acceleration boundaries .
Tlie data obtained from the literature survey and the data obtained
from the Ames investigation are presented in figures 6 through 9. For
the tolerance to acceleration tijnes obtained from the literature it
was attempted to use values wherein the subject was within about 'lO percent
of the specified acceleration value, rather than to measure the tolerance
tune irom the beginning of onset of the acceleration force to the removal
j of the acceleration force. It should be noted, however, that in many
01 the reference reports, no exact definition of tolerance time was
given and, hence, the listed tolerance tme values may have been the
total length of the run. The currently accepted boundaries defining
human tolerance to sustained acceleration for the eyeballs -out eyeballs -
down, and eyeballs -in inertial force directions are presented as dashed
lines m figures 6, 7, and 8, respectively. The data points on which
the dashed-line boundaries are based were obtained by averaging the
measured tolerance times for several test runs of nonpilot subjects.
It is felt that the dashed-line boundaries are conservative. In
contrast, the data points on which the new tolerance boundaries are
based were obtained from runs by test pilots who were preconditioned
to the effects of acceleration forces or from maximum tolerance -time
runs completed by members of a group of nonpilot test subjects . These
data points were in some cases the result of a single test run. It is
therefore anticipated that the proposed new boundaries apply only to a
lairly select group of which test pilots are members.
Eyeballs -Out Case
Figure 6 presents the available data for time tolerance to sustained
accelerations for the eyeballs-out case.
Perhaps the most consistent and complete tests on tolerance to
eyeballs -out acceleration were conducted by Clarke and Bondurant
(ref. 3). The boundary obtained from this investigation is shown by
the dashed line in figure 6. In these tests the subjects were in an
essentially normal seated position. The anterior torso and extremity
restraint system was somewhat similar to the restraint system used in
the Ames tests . The head-restraint system for the Clarke tests how-
ever, was arranged so that most of the weight of the head was taken
across the subject's forehead. It should be noted that nonpilot sub-
jects were used in this test.
The data obtained from the tests conducted in the present study
are plotted as circular test points in figure 6. In a comparison of
I tVfyZ c^nbiects in the Ames data show a tolerance tune of i^ to
7g whereas f!^,^^^^^^' increase in tolerance is attributed mainly to
IrTZrllefrilir^l^Tslel a.d the use of highly motivated test pilots
as centrifuge subjects.
The work by Balllngar and Demp^ey (ref . M i= sho™ by the t'l^^Sular J^
test ^Inirj^these^es.. he restraint ^
:S::.s were usea .„ t, -^-ser .e.t.^ho.,.er.^o^^a^^^ pe.
centage Of the ^^^^^^;^^, ^^„^° ^^^e of course, those who were most
aid pain associated with the endurance test trials.
.cce4:SS-a: r^^tlf SVa^r^ - 'rn-r|r2g-S test .
-^ "*:rtS nrrnSeSs\inh::rd'-d^rtL^?.sVy''i::^;n.^L.
llllfroi a lighted chS placed about 30 =n from the eyes A rjeasura
TJZl de^Lity was obtained by ^^^^^^^ ^^^Jt.
^'^'''ITT.^erTe I" 1 rS A f^roTofTf^r as long as 38 seconds
in a lOg field. .^^^^,^^^, ^Ji^^.f^i^^ the eyes. This reduction in
in vision which unproved after blinking ^^^ eye
visual acuity was attributed by Gauer and Ruff to the tear txui
accumulating over the lenses of the eyes.
A l?e run for 15 seconds was reported ir, reference 23- The
referencfr^^ri indLates that t^-e data « ^, i^d ^^^n^^^^
IS^r^fir:LrariarirregSi5 l^r::. SSItlons .or this program,
case the- bSor^a^TrS: "p^SSL!^ ^J -nC-Sfr ^^
T^e work conducted by Duane and others (ref . 5) showed that a pilot
• +ZrLs??ion can tolerate backward accelerations up to and
ScLri^ri/gfoiTsronL. Duane employed a restraint system of
padded 'barriers in the front of the lower face, chest, and legs. Here
again, nonpilot subjects were used, and only the hardiest of subjects
apparently completed the 15g ^nm.
The single data point shown in the impact acceleration region was
the much publicized run of Stapp (ref • 7) wherein he endured 25g
eyeballs -out force for about 1 second. It has been included in figure 6
to show the voluntary endpoint of human exposure to eyeballs -out
accelerations. Stapp was injured in this run; however, his injuries
were apparently not permanent in nature. It should be noted that Stapp' s
A head was not restrained during this run. From a pure tolerance to
k acceleration standpoint, it would appear that a healthy, high]^ motivated
3 male , as exemplified by a test pilot , can withstand acceleration fields
6 for the times indicated by the solid-line boundary in figure 6, provided
he is suitably restrained.
Eyeballs -Down Case
A procedure similar to the one outlined for the eyeballs -out
acceleration direction was also made for the eyeballs -down acceleration
direction. Figure 7 presents the available time -tolerance data for
this g field direction. For all the data points presented in this figure
the test subjects were wearing anti-g suits.
The most complete set of data on tolerance to eyeballs -down
acceleration forces was obtained by Miller, et al. (ref. 10). Nonpilot
subjects were used in this investigation- For the tolerance tests the
subjects were apparently in a normal seated position. Signal lights
were used to determine visual loss. Acceleration forces from 3 to 6g
were investigated in this research program. Exposures as long as an
hour at 3g were tolerated by the test subjects; however, these data
do not appear on figure 7 because of the limited time scale. The
dashed line in figure 7 illustrates the time tolerance to eyeballs -down
acceleration boundary derived from this set of data.
Human tolerance to 9g for 15 seconds was reported in reference 23-
There is little information available on this data point. The reference
report indicates that these data were obtained from unpublished work
conducted by the University of Southern California and that the
centrifuge subjects were wearing g protective equipment.
Acceleration force levels of 7g for 30 seconds were investigated
by Dorman, et al. (ref. 12)- In these tests the centrifuge test subjects
consisted of nonpilot laboratory personnel and active duty fleet pilots
selected at random from the operating squadrons. The test subjects
were seated in the normal position and were secured by a lap belt and
shoulder harness. Deterioration of peripheral vision was assessed by
having the subject turn off peripheral lights through a push-button
arrangement. Only 3 out of 2^ pilot subjects Euccessfully withstood
the 30 second run at 7g without anti-g suits; Irowever^ with anti-g
suits, 16 out of 2^ pilot subjects withstood tie prescribed g stress.
None of the nonpilot personnel were able to tolerate the prescribed
The triangular symbols indicate human tolerance times of about 1.2
minutes to normal acceleration values of 6.6g. These data were obtained
from unpublished centrifuge time histories obtained from the Langley
Research Center of NASA. The subjects used in the Langley tests were
experienced test pilots. For these test runs 1he pilots were seated in A
a contoured couch similar to that used in the imes tests. The pilot ^
task consisted in controlling a simulated vehicle along an atmosphere 3
entry trajectory. 6
The data obtained from the Ames tests are plotted as circles on
figure 7. The Ames data show that the test pi.\ot subjects could with-
stand 6g in an eyeballs -down direction for as :,ong as 6-I/2 minutes.
The subjects reported that at the beginning of the run there were no
physiological problems other than a momentary 1 lurrlng and dimming of
vision. As the run progressed, the pilot's vii; ion grew dimmer. During
the last I-I/2 minutes of the run the pilot in: icated he was having
considerable trouble locating the target on tht scope. The run was
terminated when the pilot could no longer tell exactly the position of
the target. Other than breathing becoming more- labored there were no
adverse physiological effects. There was no feeling of pooling of blood
in the extremities and no pain.
As can be seen there is a scarcity of dat;. on which to base any
new tolerance to acceleration boundary for the eyeballs -do-vm g field
direction. However, on the basis of the exist: ng information, a
tentative boundary has been drawn and is shown by the solid line in
figure 7. It is believed this boundary is val:.d for a test pilot subject
wearing an anti-g suit.
Eyeballs -In Case
Figure 3 presents a summary of the available data on human tolerance
to sustained accelerations for the eyeballs -in g field direction.
The dashed line boundary in the figure wa;; derived from the research
program of reference 3. It is believed the da^a from this program
represent the most complete set of results on human tolerance to this
g field direction. Nonpilot subjects were used in this experiment. Loss
of vision, inability to breathe, or pain suffi'iient to interfere with
judgment or performance were considered valid ond points to the test
run. The test subjects were positioned so tha: their legs were sharply
flexed with the trunks and heads tilted 2')° i:i the direction of the
acceleration. Reference 3 concidered this to be the position for
maximum tolerance to eyeballs -in accelerations. In this position
blackout was not observed below lOg and substernal pain vas minimum.
An average tolerance time of 5 seconds at 12g was demonstrated in this
program. It might be noted, however, that one of the test subjects
tolerated 12g for 1^1- seconds .
Reference 5 reports on a centrifuge investigation conducted by
Duane , et al. Nonpilot subjects vere seated in a standard ejection
seat from a Wavy jet fighter airplane. Conventional lap belts and
A shoulder harnesses were used to restrain the subject in the seat. The
^r task, required of the test subjects, consisted in turning off center
3 and peripheral lights through a finger switch arran.gement . In this
6 study the subjects were exposed to an acceleration force of 15g for
i; seconds . It was noted in the reference report that as soon as the g
stress was removed, the subject was not debilitated- This means that
if voluntarily or involuntarily caught in this position, a pilot could
recover instantly and perform intricate movements which might be life
saving after removal of the inertial f orce-
in unpublished work by the AMAL, RADC, Johns vi lie , Pa. , a nonpilot
test subject was immersed in an acceleration field greater than or equal
to 15g for a period of approximately l'"> seconds. The subject was
restrained by a molded couch contoured to fit the posterior shape of the
body with the subject positioned in the couch so that his upper torso
and head were held at an angle of approximately 10° with the horizontal.
Tlie knees were propped up so they were near the same level as the chest.
Ihe subjects reported blurring of vision at the higher g levels; however,
a side-arm controller could be manipulated by the test subject.
Reference 2 gives some results obtained by the investigator Buehrlen-
The subjects used in this investigation consisted mostly of jianlor
surgeons of a German military academy- The subjects were essentially in
a normal sitting position with their backs supported by an upholstered
board. In this study, peak accelerations of 17g were investigated.. The
results of the investigation showed that the subjects could withstand
10 to 12g without difficulty; however, above I'+g the subjects reported
their vision had deteriorated and they could only see dark clouds with
stars, etc. Most of the tabulated data presented in this reference
indicates only the total length of the centrifuge run and does not show
the period of time the subject was at or above a given g level. A
single time history of a tolerance run is presented in reference 2, which
shows that the test subjects were held at or in excess of I2g for 0.72
minutes. This single data point has been plotted in figirre '■:?.
Reference h reports on a series of centrifuge tests of subjects in
a semisupine position. The body was flexed at the hips so that the
head, chest, and abdomen were raised to make an angle of approximately
20° vrith the horizontal. The knees were propped up so they were at the
same level as the head.. Nonpilot subjects were used in these tests with
many of the subjects having no prior centrifuge experience. During these
test trials the subjects were required to turn off center and peripheral
lights through a three -switch arrangement situeted on a hand grip. The
subjects were also required to read word lists and perform a memory
association test. Certain subjects were able to tolerate lOg for as
long as 2 minutes. An opinion, expressed in reference k , was that
2 minutes did not represent the maximiim time tolerance to lOg.
The results of the Ames tolerance investigation are shown as circles
in figure 8. In this case the subjects tolerated 6g eyeballs -in for
approxijaately 6 minutes. It might be noted that in these tests the A
pilots were not seated in a position for maximum tolerance to eyeballs- h
in acceleration. It was surmised that had they been positioned 3
differently, their tolerance time to this magnitude and direction of 6
acceleration force would have been somewhat gre^ater.
From the data in figure 8 a new tolerance boimdary to eyeballs -in
acceleration has been drawn. It is believed that the tolerance boundary
represented by the solid line is valid for a test pilot subject suitably
restrained in a near sitting position or in a semisupine position.
The data of time tolerance to acceleration obtained in the diagonal
g field direction of eyeballs down and out is ]iresented in figure 9-
In this case it can be seen that a maximum g 1« vel of Q.k- was tolerated
for as long as 20 seconds. This g field direcl ion was particularly-
uncomfortable for the pilot because of the pain associated with blood
pooling in the extremities. A tentative boundJiry to this direction
of applied g is shown by the solid line faired through the data points .
It might be noted that no additional time tolerance data were available
for this diagonal g field direction.
In the Ames tests of tolerance to acceler;,tion, post run comments
by the test pilot subjects portray realistically the physical sensations
encountered during the test trials. These comiients are on file at the
Ames Research Center.
A summary plot showing the derived time tolerance to acceleration
boixndaries for the principal g field directioni. of eyeballs down,
eyeballs in, and eyeballs out is presented as ;'igure 10. It is well
known that the pilot cannot tolerate g forces applied in the normal
direction as well as he can tolerate g forces applied in the transverse
direction. It had been speculated by several 'jivestigators (refs. 3
and 9) that man's tolerance to eyeballs -out aci;elerations was equal to
his tolerance to eyeballs-in accelerations. Tlie results shown in fig-
ure 10 would tend to confirm these speculation:;. The tolerance boundaries
to eyeballs-in and eyeballs -out accelerations are shown as being one and
the same. One of the major physiological problems encoimtered by a person
immersed in a high acceleration field is his inability to breathe properly
(ref. !+)■ With the pilot positioned for optim^jm tolerance to the applied
acceleration force, indications are that breathing is considerably easier
during eyeballs -out thaxi during eyeballs -in accelerations. An explana-
tion for this was offered by Gauer and Ruff in reference 1 A word of
caution should be inserted here regarding the use of the derived toler-
ance boundaries . The pilot of an orbital vehicle will be in a weightless
state for extended periods of time before the entry phase of the mission.
It is speculated these extended periods of weightlessness may alter his
tolerance to high accelerations .
OSiere is a paucity of data from which to draw conclusions on man's
ability to perform a control task when he is immersed in an elevated
acceleration field. From axi extrapolation of the results of the Ames
tests and the results of other tests, it would appear that the pilot's
A ability to perform a manual control task has markedly deteriorated when
k he Is exposed to eyeballs -out or eyeballs -in accelerations greater than
3 12g. It has been stated by Duane and others (ref . 5) that, between 12g
6 and 15g, the pilot is capable of simple manual switching operations
using the hands and fingers, and the study by Clark and others (ref. 8)
has indicated that forearm, hand, finger, and ankle movements were not
impaired at 12g. Above 15g there Is the possibility of injury to the
subject and less possibility that the pilot could assume primary control
of the vehicle after removal of the acceleration stresses. In figure
10, the shaded area denotes the region of reduced pilot performance for
the eyeballs-in and eyeballs-out acceleration forces. From the results
of the Ames study and the study of reference 15, it would appear that the
pilots' vision was greying out and they were on the verge of blackout
for normal acceleration forces greater than about 6 to Tg- It is probable
that because of this visual impairment pilot control performance deterio-
rates above 6 to 7g for the normal g field direction. The shaded area
in figure 10 shows a tentative region of reduced pilot performance for
the eyeballs -down g field direction.
The dashed curve in figure 10 labeled "Entry from parabolic veloc-
ity" was computed for a drag -modulated vehicle flying along a ballistic
entry trajectory with the vehicle initial velocity taken as parabolic.
Each point of the curve represents a different atmosphere entry tra-
jectory starting from a different initial entry angle. The curve shows,
for example, that by proper drag modulation the maxim\jm acceleration
which the vehicle would encounter during an entry could be 8g and this
level of acceleration must be endixred for about 1-2/3 minutes. It has
been presumed that structures are currently available which will with-
stand the heating dictated by the entry conditions making up this
curve. On the return from a lunar mission, the depth of the entry
corridor, which must be acquired in order to effect a landing on the
earth, increases as the allowable entry accelerations increase (ref. 2^).
Thus it is desirable to enter at the high g portion of this curve, since
this reduces the accuracy demanded of the midcourse navigation and
guidance system. The conclusion is reached that for the re-entering
manned lunar vehicle^ man is still the weakest link in the chain. The
presence of man would probably prevent the vehj.cle from flying at the
sustained accelerations for which it can be mace structurally safe and
which would allow an attendant reduction in the accuracies demanded of
the navigation system.
Tlie curve for the entry from circular velc cities is presented in
figure 10 to show the maximum acceleration and length of time which must
be endured by an occupant of a drag -modulated ballistic vehicle entering
the earth's atmosphere from a circular orbit. Each point of the curve
represents a different atmosphere -entry trajectory; however, each point
of the curve is computed for an initial entry &ngle of -5° • This curve
shows the severest acceleration stress which man would probably be required
to endure on a controlled^ drag -modulated, ballistic re-entry from a
circular orbit. As can be seen from the figure , man, if properly
restrained, is apparently capable of withstanding these stresses.
EVALUATION OF SIDE-AIM CONTEC:KLERS
An additional item which can strongly infr.uence the performance
and efficiency with which a pilot can fly a veliicle in an elevated
g field is the design of the pilot's side-arm controller. In an attempt
to negate the effects of acceleration forces 02. the ability of a pilot
to control a vehicle, various side-arm controllers have been proposed.
It appears, as of the present time, that three-axis side-arm controllers
are receiving the most serious consideration. With this type, the pilot's
legs can be firmly restrained and they are not used to make control
inputs. An alternate class is the two-axis si("_e-arm controller- It is
similar to the three-axis class, except the yaw control is obtained
through movement of the feet or legs. The argimient as to which class of
controller is better hinges (l) upon whether tlie high acceleration
forces would render the legs useless for making control inputs, and (2)
upon the ability of the pilot to blend and app:y three (instead of two)
different control inputs with one hand. An additional objective of
the side-arm-controller study was to deterTnine the best side-arm
controller from configurations which represent the present state of
the art .
The procedure for evaluating the side-arm controllers was very
similar to that used in the rest of the study. To each test controller
the pilot assigned numerical ratings on vehicle controllability. After
each run, the pilot was thoroughly interrogated on the desirability of
certain controller characteristics, such as bri.^akout force, force
gradients, and axes of control rotations.
Each controller was tested in the earth's gravitational field
(static run) and in two elevated accleration f'_elds, and two to three
different sets of airframe dynamics were utili,:ed. The two elevated
test accelerations were as follows :
% = ^6; % = Og aJ^d Ax = -2g, Aj^ = hi
These accelerations were chosen as typical of those which might "be
encountered during the launch and entry phases of an orbital mission.
The vehicle longitudinal and lateral-directional airframe dynamic
characteristics^ which are shown in table III, ranged from a well-damped
system with moderate control-moment cross coupling (i.e. application
of the ailerons produced both rolling and yawing moments) to a lightly
damped system with heavy control-moment cross coupling. The parameter
A lOOC^pCng /CnpC2g_^, which Is discussed in reference 22, was used as a
measure of the control-moment cross coupling. It was believed that
3 the lightly damped heavily cross -coupled dynamic situation would empha-
size existing deficiencies in the various controller configurations.
Figure 11 shows the input axes of rotations for the various test
controllers in this investigation. The axis running parallel to the
forearm should be regarded as being essentially the center line of the
forearm. Sketch F is intended to show that the toe pedals were actuated
by differential rotation of the feet about the ankle joint. Photo-
graphs of the various test controllers and a photograph showing the
lower leg restraints and the toe-pedal installation is presented as
figure 12. The controllers were designated A, B, C, D, E, and toe
pedals. Controllers A and B were in the three-axis class. Controllers
C, D, and E were in the two-axis class. The three-axis side stick
controller A was converted into a two-axis controller by freezing the
yaw control axis. As a two-axis controller it was labeled controller C
Note that controller E is held by the fingers (fig. 11).
The force characteristics of each controller, as measured in the
earth's constant Ig field, are shown in figure 13 . The control forces
presented in this figure were measured at approximately the mid-point
of the stick grip. Fairly complete descriptions of the mechanical
features of controllers A and E are given in references 25 and 26,
respectively. No published references are available giving the design
details of the remaining side-arm controllers or toe -pedal controls;
however, the mechanical design of these latter items was reasonably
straightforward. In general, the force gradients of these controllers
were obtained by a coiled spring arrangement with a mechanical feature
which allowed some adjustment in the controller breakout forces.
When the controllers were operated in the earth's Ig field, the
consensus of the pilots was that side-arm controller, toe -pedal force
gradients, and breakout forces were acceptable for normal operation;
however, the following specific criticisms were offered:
Controller A ; The breakout force and force gradient for
the directional axis of control were higher than desired. The
roll-axis breakout force was high and the roll-axis force
gradient was too low.
Controller B ; Itie breakout forces about all axes for this
controller were high; in addition^ the roll-axis breakout forces
for right stick deflection were considerabiLy higher than those
for left stick deflection.
Controller D ; A more positive stick centering force for
the roll-axis of control was desired.
Controller E and toe pedal controls ; No specific criticism.
For the controllers used in these tests the pilot control input ,
was transmitted through a mechanical linkage to electric potentiometers.
This mechanical linkage usually consisted of a small number of links j.
with a minimum of backlash and friction at each connecting pointy with -^
the consequence, that the damping forces present in the controllers
were fairly small for some of the controllers tested. An indication of
the damping forces present in the linkage syste:a of the vario-us control-
lers was obtained by measuring the cycles to daap to half amplitude
(Ci/a) of the free oscillations about each axis of each controller.
The natural frequency in terms of the period of the free oscillation and
the damping in terms of C^/g about each axis of each controller is
presented in table IV.
It was pointed out in reference 22 that pilot opinion of the
longitudinal handling qualities of an atmosphere entry vehicle is a
function, among other things, of the gearing between the pilot's stick
and the vehicle pitch-control power (pitch-control power gradient)
expressed as (Mg &e/^p)/^%^- The value of pitcia-control power gradient
desired by the pilots is, in turn, a function o; the type of controller
(i.e., center-stick, side-arm controller, etc.) as well as a function of
the vehicle longitudinal period and damping. Tie desired values of
pitch-control power gradient for a conventional center control stick were
presented in reference 22. A brief investigation was conducted to
determine the desired values of pitch-control power gradient for side-arm
controllers D and E- These two controllers wer; chosen for this phase of
the study since they represented two distinctly different types, namely,
hand -held and finger-held. This portion of the study was conducted on
a fixed simulator in the same manner as described in reference 22- The
results of the present study are shown in figure ik . In this figure
are shown optimum regions of pitch-control power gradient for a vehicle
with high damping, 25^^ ^ 2, and for a vehicle with low damping, 2Swn ~ 0.
It is interesting to note that in this figure the hand controller
(controller D) exhibits a broad area of acceptable pitch-control power
gradients; whereas the finger-held controller (controller E) has a more
limited range of acceptable pitch-control gradients. The information
in figure 1^ was used to select the value of pitch-control power gradient
for the various controllers used in the side-ami controller evaluation
tests. The value of pitch-control power gradients used for all handgrip
side-arm controllers (i.e., controllers A, B C, and D) and for the finger-
held controller (e) is shown in figures l^(a) End iM'b), respectively.
Figure 15 is a summary plot otitalned "by averaging each pilot's
ratings on vehicle controllahility for all the acceleration fields of
this investigation and then averaging this average rating for all the
pilots (for a given set of airframe dynamics and for a specified con-
troller). Pilot comments from these tests Indicated a unanimous pref-
erence for a two-axis controller, toe-pedal combination. The difficulty
in blending and applying three control inputs through one hand was
repeatedly pointed out by the evaluation pilots; this difficulty,
however, was not reflected in the pilots' numerical ratings when they
used a controller to fly the well -damped configuration. The preference
■^ for the two-axis controllers was much stronger for controlling the
^ lightly damped configuration than for controlling the well-damped
3 dynamic one. This was verified by the pilots' numerical rating on
° vehicle controllability presented in figure I5. An approximately
1-3/^ rating point preference of the two-axis class of controllers is
Indicated for controlling the lightly damped, heavily cross -coupled
Quantitative data as well as subjective pilot comments obtained
during the tests did not indicate a clear-cut superiority of any par-
ticular two-axis controller over the others. At a roundtable discis-
sion following the tests, participants expressed a general preference
for controller E; however, this preference was not a strong one •
Arguments in favor of the finger-held controller were as follows : There
were some indications that for short-period oscillations the pilot
could control a lower level of airframe damping with this type of con-
troller as opposed to the heavier handgrip type of two-axis controllers .
Because the finger-held controller differed from the conventional center
stick (i.e., held with fingers, Inertia very low, light-force gradients,
etc.), some pilots noted that they had less tendency to handle it like
a conventional center stick and this reduced their tendency to revert
back to center-stick control patterns when faced with a "clutch" situa-
tion. The pilots noted that with the heavier controllers and in the
higher g fields, there was an apparent increase in the inertia of the
controller and hand. As a result more effort was required to deflect
the controller, and the pilots' control Inputs were smaller and were
made very cautiously; this effect was apparently reduced to some extent
when the light pencil controller was used. Arguments not in favor of
the finger-held controller were that positioning of the hand on the
controller was critical and, as a result, fore-and-aft displacement of
the hand and arm relative to the stick, due to high ±Ax accelerations,
caused some downgrading of the controller in the opinion of the pilots .
Pilots also indicated a vague feeling of the controller being somewhat
feathery, being "tender" to use, requiring no work, etc
As for the axes of control rotations for the handgrip controllers;,
the pilots expressed a unanimous preference for the roll axis of rota-
tion to be below and to run essentially parallel to the longitudinal
axis of the lower arm, and for the pitch axis of rotation to be per-
pendicular to the roll axis and to pass through the nominal wrist pivot
point. Side-arm controllers B and D exemplify the desired positioning
of the roll and pitch axes of rotation. Agreen.ent on the desirable
positioning of the yaw axes of rotation for the three-axis controllers
was not reached.
The toe pedals, used in conjunction with the two-axis controllers,
were considered quite usable. The majority of pilots who used them
stated there was no tendency toward inadvertent inputs, and good coor-
dination of the yaw input with the roll input was possible after some
practice. No marked reduction in their usefulness was noted for the
pure eyeballs -out or eyeballs -in acceleration (maximum values of a
A-j^ = -ig and A^ = 6g were tested for periods e^s long as 5 minutes ) .
For the combination eyeballs -out and eyeballs -c.own accelerations 3
(Ax ^ -5g, Ajj = 5g and A^ = -6g, Aj^ = 6g) , the usefulness of the (^
toe pedals was diminished. Blood pooling in the lower extremities
caused numbness and pain which precluded precise yaw control inputs
with the rudder pedals. Indications were that the acceleration fields
in which the toe pedals could be successfully used could be extended
appreciably if an improved lower leg g protection system were used and
if the lower leg were positioned so that its long or tibial axis was
always perpendicular to the applied acceleraticn vector.
Interrogation of the pilots after each certrifuge run indicated
that for nearly all controllers tested, there 1 as an apparent change in
friction levels, stick-force gradients, breakoi.t forces, etc, with
different levels of the impressed acceleration field. According to
pilot opinion, these stick-force changes were i;.sually to the detriment
of the controller. It appeared that the varia1:ion in stick-force
characteristics with impressed accelerations w;.s partly due to mass
imbalance of the controllers and, in part, to c eflections in the struc-
ture of the stick, which tend to bend the movable parts with an increase
in the friction levels, etc It is recognized that these changes may
also be partly imagined as a result of physiological or psychological
effects of the impressed accelerations on the iiilot. It seemed that
the controllers exhibiting the largest appareni. changes in force
characteristics were of the high inertia, high weight, bulky type which
required considerable design effort to attain ; ome semblance of mass
balance. It would seem from the experience ga:.ned in these tests that
a prime consideration in the design of controli.ers should be to keep
them light in weight with low inertia about the control axes.
The centrifuge study showed there could bi: marked decreases in
pilot tracking performance with increases in the magnitude of the
impressed accelerations. Pilot comments indicated that in order to
have the same level of control over the vehicle, an increase in the
vehicle dynamic stability is required with increases in the magnitude
of the acceleration Impressed on the pilot. It appears that a great
deal of additional research work is warranted in investigating the
effects of sustained accelerations on the pilot performance.
The study indicated quite clearly the improvement in tolerance to
acceleration times which can be realized through relatively minor
improvements in the pilot's restraint system. It would appear that with
a suitable restraint^ the pilot's tolerance to eyeballs -out accelerations
can be made equal to his tolerance to eyeballs-in accelerations. It
is suggested in this study that more meaningful tolerance to acceleration
.\ times may be obtained by using highly trained and highly motivated
'4 test subjects, as exemplified by the test pilot.
S Finally, pilot comments indicated a unanimous preference for the
two-axis class of side controller over the three-axis class. The pedal
controls used in this study resulted in effective yaw control for most
acceleration fields of this investigation.
Ames Research Center
National Aeronautics and Space Administration
Moffett Field, Calif., April 12, I96O
1- Gauer, 0., and Ruff, S. : Die Ertraglichkeitsgrenzen fur FlieMcrafte
in Richtung Riicken - Brust (The Limits of Endurability for Centrifugal
Forces in the Direction Back to Chest). Luftf ahrtmedizin, vol. 3,
no. 3, 1959, PP- 225-230.
2. Biihrlen, L. : Versuche iiber die Bedeutung der Richtung beim
Einwerken von Fleihkraften auf den menschllchen Korper (Experi-
ments of the Significance of Direction in the Case of Action of
Centrifugal Force on the Human Body). Luftf ahrtmedizin, vol. 1,
1937, pp. 307-325-
3. Clarke, Neville P., and Bondurant, Stuart: Human Tolerance to
Prolonged Forward and Backward Acceleration. WADC Tech. Rep.
h. Ballinger, E. R., and Dempsey, C A.: The Effects of Prolonged
Acceleration on the Human Body in the Prone and Supine Positions.
WADC Tech. Rep. 52-250, 1952-
5. Duane, T. D., Beckman, Edw. L. , Ziegler, J. E. , and Hunter, H. N- :
Some Observations on Human Tolerance to Accelerative Stress.
III. Human Studies of Fifteen Transverse G. Jour. Aviation
Medicine, vol. 26, no. k^ Aug. 1955, PP- 293-303.
6. Ruff, Siegfried: Brief Acceleration: Less Than One Second.
Vol- I, pt. VI-C of German Aviation Mec icine -World War II,
Dept. of the U. S. Air Force, 1950, pp- 53^-599-
7. Stapp, John P.: Effects of Mechanical Fcrce on Living Tissue.
I. Abrupt Deceleration and Windblast- Jour. Aviation Medicine,
vol. 26, no. k, Aug. 1955, PP- 268-288.
3. Clark, W. C, Henry, J. P., Greeley, P. C, and Drury, D. R.:
Studies on Flying in the Prone Position. Committee on Aviation
Med. Rep. h-66 , Hat. Res. Council, 19^4-5 ••
9. Bondurant, Stuart, Clarke, Neville P., el aL: Human Tolerance
to Some of the Accelerations Anticipated in Space Flight.
WADC Tech. Rep. 58-I56, 1958.
10. Miller, Hugh, Riley, M. B., Bondurant, S., and Hiatt, E. P. :
The Duration of Tolerance to Positive Acceleration. WADC
Tech. Rep. 58-635, Nov. 1958.
11. Edelberg, Robert, Henry James P., et al. : Comparison of Human
Tolerance to Accelerations of Slow and Rapid Onset. Jour.
Aviation Medicine, vol. 27, no. 6, Dec 1956, pp- ^82-^89.
12- Dorman, Phillip J., and Lawton, Richard !■/. : Effect of G Tolerance
on Partial Supination Combined with th(.: Anti-G Suit. Jour.
Aviation Medicine, vol. 27, no. 6, Dec 1956, pp. ^90-^96.
13. Cochran, LeRoy B. , Gard, Perry W- , and Worsworthy, Mary E. :
Variations in Human G Tolerance to Pos ".tive Acceleration.
KM 001 059.02.10, U. S. Naval School o:' Aviation Medicine, Naval
Air Station, Pensacola, Florida, 195^-
1^. Woodllng, C H., and Clark, Carl C: St' idles of Pilot Control
During Launching and Reentry of Space '''ehicles. Utilizing the
Human Centrifuge. Report No. 59-39, list. Aero. Scl., 1959-
15. Eggleston, John M. , and Cheatham, Donald C: Piloted Entries
Into the Earth's Atmosphere. IAS Paper No. 59-98, 1959-
16. Holleman, Euclid C, Armstrong, Neil A., and Andrews, William H. :
Utilization of the Pilot in the Launch and Injection of a
Multistage Orbital Vehicle. IAS Paper No. 6O-I6, Jan. I96O.
17. Crosbie, Richard J.: Cam Designing for vhe Human Centrifuge.
NADC-MA-5512, Aviation Medical Acceleration Lab., U. S. Naval
Air Development Center, Johnsville, Pa., 1955-
10. Brooks, Charles E- : Data Sensing and Recording Techniques
Estahlished for the Human Centrifuge. MADC-MA-5306, U. S. Naval
Air Development Center, Johns vi lie , Pa., 195^-
19. Smedal, Captain Harald A., USW (MC), Stinnett, Glen W. , and
Innis, Robert C: A Restraint System Enabling Pilot Control
Under Moderately High Acceleration in a Varied Acceleration
Field. NASA IN D-91, I96O.
20. Cooper, George E.: Understanding and Interpreting Pilot Opinion.
Aero. Eng. Rev., vol. I6, no. 3, Mar. 1957, PP- ^7-51, 56.
21. Sadoff , Melvin: The Effects of Longitudinal Control-System
Dynamics on Pilot Opinion and Response Characteristics as
Determined From Flight Tests and From Ground Simulator Studies.
NASA MEMO IO-I-58A, 1950-
22. Creer, Brent Y., Heinle, Donovan R. , and Wingrove , Rodney C:
Study of Stability and Control Characteristics of Atmosphere -Entry
Type Aircraft Through Use of Piloted Flight Simulators. IAS Paper
No. 59-129, Inst. Aero. Sci., 1959-
23- Lombard C F- : Human Tolerance to Forces Produced by Acceleration.
Rep. ES-21072, Douglas Aircraft Co., Feb. 27, 19^8-
2h. Chapman, Dean R. : An Analysis of the Corridor and Guidance
Requirements for Supercircular Entry Into Planetary Atmospheres.
NASA TR R-55 , 1959 •
25. Andrews, William H., and Holleman Euclid C: Experience With a
Three-Axis Side-Located Controller During a Static and Centrifuge
Simulation of the Launch of a Multistage Vehicle. NASA TN D-5^6,
26. Sjoberg, S. A., Russell, Walter R. , and Alford, William L- :
Flight Investigation of a Small Side-Located Control Stick Used
With Electronic Control Systems in a Fighter Airplane.
NACA RM L56L28a, 1957-
TABLE I.- VEHICLE DYNAMIC CHARACTERISTICS ?0R PERFORMANCE TESTS
Vehicle dynamic parameters
Dutch roll damping ratio
Dutch roll period; sec
Roll time constant^ sec
Cross -Gouplinc parameter ^
Loncitudinal damping ratio
Lon^;itudinal period ^ sec
Indicates vehicle dynajnic characteristics for tolerance to acceleration
TABLE II.- PILOT OPINION RATING SYSTEM FDR UNIVERSAL USE
4 Acceptable, but with unpleascnt
5 unocceptable for normal
6 Acceptable for ennergency
Unacceptable even for emergency
8 unacceptable - dangerous
9 unacceptable - uncontrol lable
Motions possibly violent
enough to prevent pilot
I Excellent, includes optimum Yes
Satisfactory 2 Good, pleasant to fly Yes
3 Satisfactory, but with some
mildly unpleasant characterist cs Yes
♦(Failure jf a stability ougmenter)
TABLE III.- VEHICLE DYNAMIC CHARACTERISTICS FOR SIDE-ARM CONTROLLER
Vehicle dynamic parajneters
Dutch roll damping ratio
Dutch roll period, sec
Roll time constant, sec
Longitudinal damping ratio
Longitudinal period, sec
TABLE IV.- PERIOD AI© C
OF FREE OSCILLATION
Fig-ure 1.- Pilot's restraint system.
Figure 2.- Pilot's instrument display.
•VJI .KJC .UT.Vm/W .1
• v/ I
Figure 6.- Summary of tolerance to eyetalls-out acceleration.
D Rtl- lU
.01 .02 .04.06 .1 .2 .4 .6 I 2 4 6 10 20 40
Figure 7.- Siunmarv of tolerance to evebal] s-down acceleration.
Q REF 2
^ REF 4
.01 .02 .04.06 .1 .2 .4 .6 I 2 4 6 10 20 40
Figure 8.- Suramary of tolerance to eyeballs-in acceleration.
1 ' I I I M i l 1 I 1 I I I III I I I I I ilil I I
.01 .02 .04.06 .1 .2 .4 .6 I 2 4 6 10 20 40
Figure 9.- Summary of tolerance to eyelDalls-down and -out accelerat
01 .02 .04.06 .1
.2 4 .6 I 2
4 6 10 20 40
Figure 10.- Time tolerance to acceleration toundari
Fisure 11 . -
D TOE PEDALS
Axes of rotation for test controllers.
CD 8 -PITCH
-i4 - AXIS
m 4 -YAW
^_ 2 -AXIS
Ll 4L I I
30 20 10 10 20 30
CONTROL DEFLECTION, DEG
CONTROLLER A AND C
m 8 -PITCH
°= 4 -
tt: 2 -
u. 4 L I
-J 8 -YAW
2 4 -AXIS
o 8L I r III
^ 30 20 10 10 20 30
CONTROL DEFLECTION , DEG
oc 2 -
i^ 4i_«^T I I I
30 20 10 10 20 30
(b) CONTROL DEFLECTION , DEG
30 20 10 10 20 30
CONTROL DEFLECTION, DEG
Figure 13.- Control force characteristics,
10 10 20
CONTROL DEFLECTION, DEG
Figirre 13.- Concluded.
OPTIMUM REGION FOR
HIGH DAMPING / ^SED IN SIDE-ARM
// y CONTROLLER
. /NOTE: DISTANCE
/ FROM PIVOT POINT
^ TO CENTER OF
HANDGRIP 3.5 IN.
CPTIMUM REGION FOR
- I I '^ '0? 2
(LONGITUDINAL NATURAL FREQUENCY)? Wp^ ^^^ ^EC
(a) CONTROLLER D
Figure 1^^.- Optimum pitch contrcl power gradient.
OPTIMUM REGION FOR HIGH DAMPING
2- \ 28<On»2
^ USED IN SIDE-ARM
FROM PIVOT POINT
TO CENTER OF
KNOB 2.75 IN.
OPTIMUM REGION FOR LOW DAMPING
(LONGITUDINAL NATURAL FREQUENCY)^ con^ PER SEC^
(b) CONTROLLER E
Figiire ik.- Concluded.
o LIGHTLY DAMPED, HEAVILY
• WELL-DAMPED. MODERATELY
2-AXIS WITH TOE PEDALS
Figure 15.- Pilot rating on vehicle controllability using test
NASA - Langley Field, Va. A-^36
u an -
m 00 CQ
3 >. to
4.* CQ O CO
B H K ■
S5 S5 o • • - '^
15: S a> hg
"3 v C
B 3 C
« <S O
S rt 5
5q ^ rt _g» «
ri *S u (U rt
0) to Qj -*-•
T< ;:^ 0) to •
« o V
1 §"3 -2 ».
3 "* S B CQ Q
< M S "B H
3 V JO
TS 5 ■« '^
CO ^ QJ a)
Q, -t^ ■«-•
> ■§£ S3
» 01 3
. ja ■=• o
I <<> S
■a Q. S? «
-« o »> *.
S 5 o *-
a-c " s
« ™ 2 •-<
h o -a o- 2
4) o e _
bS "U i
■ag 3 ■* 5
ti i-i on ij
•S u <u cd d
'^ '^ "3
^ oj A
«' s ?
a » 2
8 0, B
DO 0) ■*-•
Z> O ^ >
s a §>2
■w n o CO
3 >^ a
;a o •
^ lU CQ
Oin CO eo
■<! ? a! Z
-M, TO HH V^ C--
•o . O
H c» 33
■ h u
- o S
■§ ii i i
<l> B ^
«. O « ~
0) U ^
- C3 ci S »<
i g S
S B i3 B e
B 5 " tJ S
■^ - - (H 3
B u ">
C S Cll 41
cd 0) o
o a> 5
B a s ^
o a 8 o B
rt C "
<»H u (U ed ed
fc. 1-1 on i
„ a) m •
"i 2 g
£:; * "
v o ai ^
a) a *^ Q
i-i aa -
*« a is ^
■ B ») £
Om CO eo
6 WO •
a> O pn ■
_ ►■ B M